I use quantitative models and computer simulations to study a variety of problems related to cultural transmission in human societies, and generative, evolutionary explanations for social phenomena. In particular, I am interested in methodological issues that arise when relating such models to the types of evidence available in long-term evidentiary records: archaeology, economic history, and human genetics. Most of my work at present breaks down into two categories:

1. Signaling theory and multi-level selection models as explanations for conspicuous consumption and other "runaway" status displays in human societies. A particular focus of effort at the moment is working with Terry Hunt and Carl Lipo to understand Rapa Nui (Easter Island) statue-making as an example of such phenomena.

2. Developing methods for recognizing various "modes" of cultural transmission (e.g., imitation of random individuals, prestige-biased imitation, and frequency-dependent bias) in real-world data: archaeological assemblages, textual and other historical data sets. In specific, I am focused on developing agent-based simulations to examine these "modes" of transmission within realistic spatial and network models for purposes of extending archaeological seriation and other observational techniques. Much of this work is in collaboration with Alex Bentley and Carl Lipo.

In addition to these primary research topics, I maintain a strong interest in the conceptual foundations of evolutionary biology, particularly as applied to human cultural phenomena, and the philosophies of biology, mind, language, and the social sciences.

April 3, 2008: Version 1.8.1 of TransmissionLab is available in source code form at the Google Code repository. This version includes social network graphs as population structures, additional statistics, and more flexible rules and other underlying structures. This version was used for our recent poster at the Society for American Archaeology meetings in Vancouver, B.C.

December 1, 2007: Version 1.5pre4 of TransmissionLab is available in source code form, at the Google Code repository. This pre-release of 1.5, which is not yet finalized, incorporates trait residence time statistics and graphs, and allows output of individual trait frequency snapshots at defined intervals, to facilitate a variety of analyses. The next development cycle will incorporate graph structuring for populations, and should be available by early Q1 2008.

April 4, 2007: Version 1.4 of TransmissionLab is available, in binary and source form, at the Google Code repository. This version abstracts away the Repast "model" infrastructure and allows multiple model classes to easily share a common library of agent population "factories," data collector classes, and population "transformation" (e.g., mutation, copying) rules. This "component" approach is designed to create the flexibility of component-oriented simulation platforms like NetLogo, but retain the full power and "white box" character of a low-level framework like RepastJ.

March 9, 2007: Version 1.3 of the RandomCopyModel is available, fully modularizing the mutation and transmission rule classes, but still duplicating only the functionality found in Bentley et al. 2007. Future work on CT simulation models will be concentrated in the TransmissionLab model, now hosted in a public Subversion repository on Google Code.